Clay exec slashes Slack noise 80% with AI-powered inbox system

The Gist

• Yash Tekriwal built an AI system that cuts 150 daily Slack notifications to 30 actionable ones
• Uses Perplexity Computer's multi-model orchestration to auto-categorize messages
• OpenClaw-built Slack digest runs deterministic code after initial AI setup

• Yash Tekriwal built an AI-powered Slack inbox that reduced his daily notifications from 150 to 30 by categorizing and prioritizing messages.
• Use AI to build deterministic tools rather than just performing tasks, enabling automation of structured data handling via APIs.
• Perplexity Computer’s multi-model orchestration automates subtasks with different models, eliminating manual back-and-forth.
• The anti-to-do list framework suggests dedicating time daily to automate repetitive tasks you despise.
• SaaS is not dying but will explode into specialized micro-software serving narrow use cases.
• The future of productivity lies in personalized software that extends existing SaaS platforms to fit individual workflows.

• Implement an AI-powered categorization system for Slack notifications to reduce noise and prioritize actionable items.
• Use AI to build deterministic tools (e.g., via APIs) rather than relying on repetitive manual tasks.
• Adopt the anti-to-do list framework to systematically automate disliked tasks.
• Explore multi-model AI tools like Perplexity Computer for automated, cloud-deployed workflows.

• 80% reduction in Slack noise (Yash Tekriwal's AI-powered inbox system reduced his daily Slack notifications from 150 to 30.)
• 100-150 daily Slack notifications (Yash receives this volume of notifications, but only 30-40 require real action.)